A Reexamination of Firm Size, Book-to-Market and Earnings- Price in the Cross-Section of Expected Stock Returns

Abstract:     
This paper reexamines the explanatory power of beta, firm size, book-to-market equity, and earnings-price ratio to average stock returns with correcting two currently controversial biases: the selection bias in COMPUSTAT and the errors-in-variables (EIV) problem. The selection bias is corrected by collecting most of the missing book equity data on COMPUSTAT from Moody's Manuals, and the EIV bias is corrected by using the consistent maximum likelihood estimation of risk premia. After filling in the missing data with the recovered Moody's sample, I do not find any significantly different results for book-to-market equity from using the COMPUSTAT sample only. With correcting for the EIV bias, I find stronger support for the beta pricing theory than previous studies. Regardless of the presence of firm size, book-to-market equity, and earnings-price ratio, betas have economically and statistically significant explanatory power to average stock returns. In particular, firm size is weakly statistically significant using monthly returns, but is no longer significant using quarterly returns. Earnings-price ratio is also not significant. Book- to-market equity has significant explanatory power to average stock returns, even though the EIV bias is corrected. However, a substantial portion of this significance of book-to-market equity is due to the embedded firm size. After disentangling the separate effects of book- to-market and firm size, I find that the significance of book-to-market equity is drastically decreased.

JEL Classifications: G12

Working Paper Series